Winners by victory condition

[Renata]

A modification of the current scoring system would be best, yes!

 

[DaveMcW]

I really don't like factoring population into the score of a spaceship game.

 

[Memphus]

Now this may be way to far out there, But I don't think it would be that hard to develop a different scoring system for each victory condition:



Cultural:your total empire cultural rate (in growth per turn)/total cities

Spaceship: Your total empire production per turn/ total number of cities

Domination: Total population / World population, or/and
Total population / total number of cities

Conquest: Total units / total number of cities

Diplomatic: Total votes / Total number of cities world wide or/and
Total votes / total number of civs left in the game

Time: Factor in all of the above(for diplomatic a UN vote coudl take place on the last turn to vote for the winner and the poll still be counted even if lost) but divide by 5, and without the time bonus this could be close

while keeping the early the win the better


Now any or all of these can be moddified or added onto existing/other solutions, but IMO different scoring for different victories makes sense.

 

[DaveMcW]

Here's one in the spirit of Civ4's exponential finish bonus.

GPR = 100 * 2^( 5 * (best_turn - finish_turn) / max_turns )

If we apply this formula to GOTM1 diplomatic victory, we get:

 

[DaviddesJ]

deleted post

 

[DaveMcW]

Oops, let me redo that.

 

[kingjoshi]

Well, everyone values different things and we play for different reasons, so what exactly should the score reflect?

For example, earliest victory seems to be something most people consider important, but let's say if you micromanaged your cities to finish the spaceship few turns earlier, but it would've taken more real time (micro-managing and pressing enter until launch vs just repeatedly pressing enter until launch), what's more valuable?

When you know you're going to win and how and then it just becomes grunt work (which is what milking is for me, though I can see others obviously enjoy it), why care about your score or finishing a few turns faster?

I mean, I find Dar's 2 hour victory much more impressive then anything else because saving real time allows for more games, try more things, real time to do other things, etc.

I like that Memphus has posted alternatives and I think that's a start to see what people value and think. However, how about if you're dominant (but barely for victory) AND diplomatic (but barely liked by enough). Whichever victory you take will not net you a good score, which forces people to play "Ace of one" over "Jack of all" game (assuming you want a good score). Which is more important (one Ace or many Jacks)?

If two people (A and B) both get cultural victories on the same turn, but A has many civs liking them while B is hated and A is close to a diplomatic victory as well, did A have a better game? How about if the person playing B prefers to fight and so gained more enemies? I don't know if it's possible to balance so much in the scoring, but it'll be interesting.

-----

I think DaveMcW mentioned this, but I'm not sure how similar my idea is. It would be much simpler to take the fastest finish for each VC, but instead of using the in-game scoring, curve by game turn. I don't think this deals with scoring, but is in regards to rankings.

Here's the example and idea:
100 submissions

15 domination victories
15 spaceship victories
15 diplomatic victories
15 conquest victories
15 culture victories
15 Time victories
10 losses

For each VC, we also include the losses so no victor gets a 0 ranking.
To be able to sort by turn using winners and losers, we double the number of turns and do a flip for the losers.

For example, let's say we're playing on "Example" speed and there are 500 turns. Someone winning on turn 200 would earn that number, 200. Someone winning on turn 140 would earn the number 140.

However, someone losing on turn 300 would earn the number 700 (2xTotalNumberOfTurns-TurnYouLost). So someone losing on turn 1 would get 999.

Ranking example using Domination victories.

Sort those 15 wins AND the 10 losses by earned number.
The fastest victor scores 100%. earliest loser gets 0%.
Give each other person an appropriate ranking depending upon the turn they won/lost inbetween the range.

We can do that for each VC.

----
We could also do the same for fastest real time finishes as well! Once you put the data in a DB, the function to do these would be similar and easy to account for VCs and game-turn versus real time or even in-game scoring or another score people come up with.

 

[kingjoshi]

Man, I spent a long time writing and formulating my idea that I didn't even know Dave posted again on this page...

 

[llib_rm]

There should be a way to equate culture points with beakers in some normalization function to level out the scores between spaceship and culture wins.